Kinetic model-based factor analysis of dynamic sequences for 82-rubidium cardiac positron emission tomography

Area of Science:

• Nuclear Medicine
• Cardiovascular Imaging
• Physiological Modeling

Background:

• Dynamic cardiac Positron Emission Tomography (PET) imaging is crucial for quantifying physiological function.
• Traditional factor analysis methods decompose PET images into tissue types based on temporal signatures.
• Improving the accuracy and precision of these decompositions, particularly for myocardial blood flow (MBF), remains an active area of research.

Purpose of the Study:

• To introduce and evaluate a novel kinetic model-based (MB) factor analysis method for dynamic cardiac PET image decomposition.
• To assess the physiological accuracy of the MB method using simulated and experimental (82)Rubidium ((82)Rb) cardiac PET data.
• To evaluate the precision of myocardial blood flow (MBF) measurements obtained with the MB method.

Main Methods:

• Employed a gamma-variate model for (82)Rb transport and a one-compartment model for blood-myocardium exchange.
• Conducted simulations of canine and rat heart imaging to assess parameter estimation errors.
• Validated factor and structure accuracy using arterial blood sampling in rats and (11)Carbon monoxide ((11)CO) blood pool imaging in dogs, comparing results to a minimal structure overlap (MSO) method.

Main Results:

• The MB method demonstrated significantly lower root-mean-square error (RMSE) than MSO for factor and structure estimations in both canine and rat simulations.
• MB showed improved accuracy in blood factor estimation compared to MSO in rat studies and comparable blood structure accuracy in dog studies.
• Myocardial and blood structures were more reproducible with MB, and MBF values tended to be more reproducible, with significantly shorter execution times.

Conclusions:

• Kinetic model-based factor analysis provides a physiologically accurate decomposition of (82)Rb dynamic cardiac PET images.
• The MB method holds potential for enhancing the precision of myocardial blood flow (MBF) quantification.
• The MB approach offers advantages in accuracy, reproducibility, and computational efficiency over previous methods.